An analysis of the Cumbria floods

Barely six years after record-breaking rainfall over the Cumbria fells, the north-west area of England has again been subjected to extreme rainfall that has led to widespread flooding, with all the associated disruption and misery that this causes. This time around the physical extent of the flood-affected areas, covering Carlisle, Appleby, Lancaster and the Scottish Borders, would seem to confirm that this has been a very significant flood event. Whilst the improved management measures, introduced after the summer 2007 floods, meant that there was good forecasting of the storm hitting the north-west, with the Environment Agency, local authorities and emergency services ready to respond, the immediate aftermath has seen a series of questions raised over the impact of the storm and whether more could have been done. 

The 2009 and 2015 floods compared

An intense rainfall event hit Cumbria in late November 2009, causing a flash flood to run through the towns of Cockermouth and Workington, with the loss of several bridges and the death of a policeman on duty. At Seathwaite the depth of rainfall over a 24 hour period set a new record for the United Kingdom of 316.4 mm, and subsequent reviews indicated that the likely return period was in excess of 1000 years.

The initial data from this weekend’s storm shows that the 2009 record has been broken with 341.4 mm being recorded at the nearby Honister Pass rain gauge. It will be sometime before a detailed analysis is completed, but it seems certain that these two events, whilst of exceptional depths of rainfall, will not be viewed as being so unlikely in the future.

The challenge of analysing extreme rainfall events

Estimates of extreme rainfall or river flow are important for the planning, design and operation of flood defences. But, by their very nature, data on rare events is very limited. This paucity of observed flood flow is a major challenge faced by practitioners when estimating the probability of extreme events.  One of the  limitations of current approaches, (such as the Flood Estimation Handbook), is that it is currently challenging to include sources of historical information such as flood marks from several centuries ago, anecdotal evidence from eye witnesses, descriptions from historical archives in order to reduce the uncertainty in estimates of extreme water levels.

For example, in many river basins there is often a range of incomplete or ambiguous historical data available relating to extreme flood events. However, owing to the high uncertainty related to these data they are rarely, if ever, used by practitioners to improve the confidence in their estimates of extreme flood flow, because these observations are often considered to be too complex to be easily handled.

HR Wallingford has been carrying out research using advanced statistical techniques to demonstrate how practitioners can use these novel sources of historical information to produce more accurate estimates of extreme flows and water levels. Case studies carried out in catchments in the UK and overseas show that these techniques can significantly reduce uncertainties in the estimates of extremes. This reduces the uncertainties in the design of flood mitigation measures potentially leading to cost savings and a better understanding of their performance during extreme events.

The impact of climate change?

In recent years there have been a number of flood events where very high flood flows have occurred, sometimes covering large areas and affecting major river basins (for example, the Severn and Thames in July 2007) and, at the other extreme, causing localised but severe flooding (such as at Boscastle in 2004). The 2015 flood in Cumbria was caused by very heavy rainfall over a large area, and records for the highest rainfall totals over 24 hour and 48 hour periods were broken.    

Scientists are cautious about linking the apparent increase in extreme rainfall events with climate change. However there has been an increase in storminess in recent years and many scientists now accept that climate change is a contributory factor. Advances in climate and hydrological modelling mean that it is now possible to estimate the contribution made by climate change to flood events. In an article in the Guardian on the 11 December, it was reported that researchers at Oxford University and the Royal Netherlands Meteorological Institute (KNMI) had calculated that climate change had made the flooding event 40 per cent more likely, with the estimate of the increased likelihood ranging between 5 and 80 per cent.

Flood risk management options

As a result of the 2009 flood, a series of flood defences were designed and constructed across Cumbria at a total cost of £45 million, to protect the most vulnerable areas and greatest number of properties. The UK Government reported that defences protected over 8000 homes in the recent flood. However, the December 2015 flooding in Cumbria has affected several areas where flood defences already exist.  The magnitude of the floods was so great that the existing defences were overtopped. These defences would have been designed for a particular magnitude of flood, for example, a flood event with an expected annual probability of one per cent (the ‘1 in 100-year’ flood). However these ‘design events’ are likely to occur more frequently in the future, and events that overtop the defences will also be more frequent.

The implications for flood risk management are that different approaches will be needed to reduce flood risk. The traditional approach in many areas is to construct flood walls and embankments in order to contain the flood water.  However, as flood flows increase, the flood water levels will increase and therefore the crest levels of the walls and banks would have to be raised.  This would result in higher and higher defences which, should they overtop or fail, could lead to catastrophic flooding.  

Therefore, rather than construct higher defences, an alternative approach would be to adopt measures that reduce the river flood flows and/or flood levels. The types of measures that could reduce the river level include setting back the defences to allow more space for the flood water, removing obstructions to flood flows, enlarging the river channel, lowering the floodplain level, dredging and the construction of flood relief channels. Measures that reduce the river flow include the storage of flood water upstream in polders on floodplains or reservoirs, or by the use of catchment water retention.  

Conclusion

The events in Cumbria during December 2015 are a reminder that even with the best plans and preparations, nature can still throw storm events at us that we are not able to deal with fully. This is the philosophy of flood risk management, that we plan and invest for a certain range of storms, but we cannot cover every eventuality. 

We have certainly made progress over the past 15 years in the UK, and we continue to improve our understanding and responses to major flooding events. Given that floods come in all sorts of forms and sizes, it is right a portfolio of measures is used to deal with them. This includes strategic flood defences to protect against the major floods from our rivers and coasts, property-level barriers and doors for those households at risk of surface flooding or as a secondary barrier in the most vulnerable areas. A new British Standard on flood resilient construction methods was produced in November 2015, with the input of HR Wallingford, which can help in reducing the impact of flood water on properties, saving on cleaning and repair costs, and enabling families to return home in a shorter time frame.

Improving the communication of flood risk and resilience options is vital, and the Environment Agency and National Flood Forum are tackling these challenges head on.